Synchronizing tasks - multithreading

Question

I am develeping an application that uses a threadpool, submits tasks to it and synchronizes them. The main thread has to wait until all the submitted tasks from a single loop iteration finish and then it submits another bunch of tasks (because the tasks from the next iteration operate on the same data and they will be dependent on one another).
My question is, what is the best way to do that?
So far, what I have come up with is that each thread, after finishing a task, increments an atomic unsigned integer. When the integer equals the number of submitted tasks, the main thread continues its work and submits another round of tasks.
This is my first multithreaded application.
Is this an optimal and sensible way of dealing with this problem.
I'm using a threadpool class copied from an excellent book "C++ Concurrency in Action: by Anthony Williams.
Here are the classes:
class thread_pool
{
std::atomic_bool done;
thread_safe_queue<std::function<void()> > work_queue;
std::vector<std::thread> threads;
join_threads joiner;
void worker_thread()
{
while(!done)
{
std::function<void()> task;
if(work_queue.try_pop(task))
{
task();
}
else
{
std::this_thread::yield();
}
}
}
public:
thread_pool():
done(false),joiner(threads)
{
unsigned const thread_count=std::thread::hardware_concurrency();
try
{
for(unsigned i=0;i<thread_count;++i)
{
threads.push_back(
std::thread(&thread_pool::worker_thread,this));
}
}
catch(...)
{
done=true;
throw;
}
}
~thread_pool()
{
done=true;
}
template<typename FunctionType>
void submit(FunctionType f)
{
work_queue.push(std::function<void()>(f));
}
};
template<typename T>
class threadsafe_queue
{
private:
mutable std::mutex mut;
std::queue<T> data_queue;
std::condition_variable data_cond;
public:
threadsafe_queue()
{}
void push(T new_value)
{
std::lock_guard<std::mutex> lk(mut);
data_queue.push(std::move(new_value));
data_cond.notify_one();
}
void wait_and_pop(T& value)
{
std::unique_lock<std::mutex> lk(mut);
data_cond.wait(lk, [this]{return !data_queue.empty(); });
value = std::move(data_queue.front());
data_queue.pop();
}
std::shared_ptr<T> wait_and_pop()
{
std::unique_lock<std::mutex> lk(mut);
data_cond.wait(lk, [this]{return !data_queue.empty(); });
std::shared_ptr<T> res(
std::make_shared<T>(std::move(data_queue.front())));
data_queue.pop();
return res;
}
bool try_pop(T& value)
{
std::lock_guard<std::mutex> lk(mut);
if (data_queue.empty())
return false;
value = std::move(data_queue.front());
data_queue.pop();
}
std::shared_ptr<T> try_pop()
{
std::lock_guard<std::mutex> lk(mut);
if (data_queue.empty())
return std::shared_ptr<T>();
std::shared_ptr<T> res(
std::make_shared<T>(std::move(data_queue.front())));
data_queue.pop();
return res;
}
bool empty() const
{
std::lock_guard<std::mutex> lk(mut);
return data_queue.empty();
}
};
The main() function:
std::condition_variable waitForThreads;
std::mutex mut;
std::atomic<unsigned> doneCount = 0;
unsigned threadCount = 4; // sample concurrent thread count that I use for testing
void synchronizeWork()
{
doneCount++;
if (doneCount.load() == threadCount)
{
doneCount = 0;
std::lock_guard<std::mutex> lock(mut);
waitForThreads.notify_one();
}
}
void Task_A()
{
std::cout << "Task A, thread id: " << std::this_thread::get_id() << std::endl;
std::this_thread::sleep_for(std::chrono::milliseconds(3000));
synchronizeWork();
}
int main()
{
unsigned const thread_count = std::thread::hardware_concurrency();
thread_pool threadPool;
for (int i = 0; i < 1000; ++i)
{
for (unsigned j = 0; j < thread_count; j++)
threadPool.submit(Task_A);
// Below is my way of synchronizing the tasks
{
std::unique_lock<std::mutex> lock(mut);
waitForThreads.wait(lock);
}
}

Answer 1

I am not familiar with the threadpool class you are using.
Without using such a class, the usual way to do this looks like this:
std::cout << "Spawning 3 threads...\n";
std::thread t1 (pause_thread,1);
std::thread t2 (pause_thread,2);
std::thread t3 (pause_thread,3);
std::cout << "Done spawning threads. Now waiting for them to join:\n";
t1.join();
t2.join();
t3.join();
std::cout << "All threads joined!\n";
I would imagine that any decent threadpool class would allow you to the same sort of thing, even more simply, by giving you a metho to block until all the threads have completed. I suggest you double check the documentation.